Author Topic: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap  (Read 27237 times)

0 Members and 1 Guest are viewing this topic.

Offline plesa

  • Frequent Contributor
  • **
  • Posts: 965
  • Country: se
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #25 on: August 15, 2014, 06:19:18 pm »
I wish my 'scope (DSO-X 2024A) had switchable input Z (50/75/1M) as I do loads of RF and video work.
I have to use T pieces and terms - very messy - easy to use the wrong term.

Using a spectrum analyzer and a tracking generator is a great way to show the effect of a stub.
Try to ask in the MSOX/DSOX 2k/3k hack thread, there is several options which can be enabled in firmware. I'm also not sure if the HW (relay) is present.
But 75 Ohm is rare on scopes nor is it available on 3K/4k/6k.
 

Online PA0PBZ

  • Super Contributor
  • ***
  • Posts: 4601
  • Country: nl
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #26 on: August 15, 2014, 06:25:17 pm »
I wonder if in the last part where the yellow wave peaks when the blue one is at its lowest... is this maybe the function generator increasing the voltage? If it tries to keep a constant level by measuring the output it will certainly be influenced by the reflected signal.
Keyboard error: Press F1 to continue.
 

Offline briandorey

  • Regular Contributor
  • *
  • Posts: 51
  • Country: gb
    • Project Blog
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #27 on: August 15, 2014, 07:54:43 pm »
You can also use the stub as a notch filter to attenuate certain frequencies which can be useful to block radio receivers from strong interference sources.
I found this site http://www.arcticpeak.com/antennapages/quaterwavestub.htm which has a useful calculator to find the coax length needed to block a frequency.
 

Offline HP-ILnerd

  • Frequent Contributor
  • **
  • Posts: 259
  • Country: us
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #28 on: August 15, 2014, 09:33:45 pm »

That AT&T video is one of my favorites - absolutely brilliant way to visualize wave propagation and reflections on a transmission line.

Yeah, and I like the broad applicability of it .  You'll never look at a wave in a pool in quite the same way again.

It's a very clear and watchable video, too.  As are yours!  Between them, a beginner's train of thought can go from "what the heck is going on?" to "that needs to be terminated."

Understanding is powerful stuff.
 

Offline VanitarNordic

  • Frequent Contributor
  • **
  • Posts: 277
  • Country: 00
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #29 on: August 16, 2014, 03:20:41 pm »
Please also make a video on differential signal measurement (like CAN ...)
 

Offline SNGLinks

  • Regular Contributor
  • *
  • Posts: 136
  • Country: gb
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #30 on: August 16, 2014, 10:02:14 pm »
I wish my 'scope (DSO-X 2024A) had switchable input Z (50/75/1M) as I do loads of RF and video work.
I have to use T pieces and terms - very messy - easy to use the wrong term.

Using a spectrum analyzer and a tracking generator is a great way to show the effect of a stub.
Try to ask in the MSOX/DSOX 2k/3k hack thread, there is several options which can be enabled in firmware. I'm also not sure if the HW (relay) is present.
But 75 Ohm is rare on scopes nor is it available on 3K/4k/6k.

I've got most of the options enabled. Never heard of different terms being an option. As the 'scope has many video trigger modes I'm surprised it doesn't offer the video term. 
 

Offline OilsFan

  • Regular Contributor
  • *
  • Posts: 52
  • Country: us
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #31 on: August 17, 2014, 02:08:03 am »
Thanks HP-ILnerd and w2aew for the videos. Very informative.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 6388
  • Country: gb
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #32 on: August 17, 2014, 05:55:55 am »
I wish my 'scope (DSO-X 2024A) had switchable input Z (50/75/1M) as I do loads of RF and video work.
I have to use T pieces and terms - very messy - easy to use the wrong term.

Using a spectrum analyzer and a tracking generator is a great way to show the effect of a stub.
Why would anyone put a 75 ohm terminator on a 50 ohm connector. There are 75 ohm BNCs, although they are rare these days. Beware of these things, as they cross-mate mate too easily with a 50 ohm BNC, and bust it.

Adding 50 ohm termination is not cheap if you do it right. You need a switch, and it probably needs to be a physical switch. If you switch in a 50 ohm terminator while the maximum permitted voltage is applied to the input there would be considerable power dissipated, so you also need protection too.

When the input switching of scopes was largely mechanical a 50 ohm terminator setting was much more common. Quite a few scope didn't have proper protection, though, and dealing with overdissipation in the terminator was a common reason for repairs.
 

Offline SNGLinks

  • Regular Contributor
  • *
  • Posts: 136
  • Country: gb
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #33 on: August 17, 2014, 11:24:13 am »
I wish my 'scope (DSO-X 2024A) had switchable input Z (50/75/1M) as I do loads of RF and video work.
I have to use T pieces and terms - very messy - easy to use the wrong term.

Using a spectrum analyzer and a tracking generator is a great way to show the effect of a stub.
Why would anyone put a 75 ohm terminator on a 50 ohm connector. There are 75 ohm BNCs, although they are rare these days. Beware of these things, as they cross-mate mate too easily with a 50 ohm BNC, and bust it.

Adding 50 ohm termination is not cheap if you do it right. You need a switch, and it probably needs to be a physical switch. If you switch in a 50 ohm terminator while the maximum permitted voltage is applied to the input there would be considerable power dissipated, so you also need protection too.

When the input switching of scopes was largely mechanical a 50 ohm terminator setting was much more common. Quite a few scope didn't have proper protection, though, and dealing with overdissipation in the terminator was a common reason for repairs.

When working with video you have to use a 75 ohm term. I use 75 ohm cable and a 75 ohm term plugged into a 50 ohm T-piece into the 'scope. The T-piece doesn't appear to suffered having the wrong BNC plugged into it. I would not risk a 75 ohm BNC plugged directly into the 'scope.

If you weren't meant to look at video then why give us all those video triggering modes?
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 6388
  • Country: gb
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #34 on: August 17, 2014, 03:35:28 pm »
I wish my 'scope (DSO-X 2024A) had switchable input Z (50/75/1M) as I do loads of RF and video work.
I have to use T pieces and terms - very messy - easy to use the wrong term.

Using a spectrum analyzer and a tracking generator is a great way to show the effect of a stub.
Why would anyone put a 75 ohm terminator on a 50 ohm connector. There are 75 ohm BNCs, although they are rare these days. Beware of these things, as they cross-mate mate too easily with a 50 ohm BNC, and bust it.

Adding 50 ohm termination is not cheap if you do it right. You need a switch, and it probably needs to be a physical switch. If you switch in a 50 ohm terminator while the maximum permitted voltage is applied to the input there would be considerable power dissipated, so you also need protection too.

When the input switching of scopes was largely mechanical a 50 ohm terminator setting was much more common. Quite a few scope didn't have proper protection, though, and dealing with overdissipation in the terminator was a common reason for repairs.

When working with video you have to use a 75 ohm term. I use 75 ohm cable and a 75 ohm term plugged into a 50 ohm T-piece into the 'scope. The T-piece doesn't appear to suffered having the wrong BNC plugged into it. I would not risk a 75 ohm BNC plugged directly into the 'scope.

If you weren't meant to look at video then why give us all those video triggering modes?
Oscilloscopes are meant for the viewing of many types of signal, but the oscilloscope doesn't provide every possible interface needed for that.

If you have a real 75 ohm BNC it has a thinner pin in the middle, so its impedance is higher. They easily mate with a 50 ohm BNC, but the narrower socket will be stretched by the thicker pin. I have no idea when they didn't make the outside part of the 75 ohm version fatter to get the right impedance in a way that wouldn't allow false mating.

These days its common to find a 50 ohm BNC on the end of 75 ohm cables. YMMV.
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 5958
  • Country: au
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #35 on: August 19, 2014, 05:12:18 am »
I wish my 'scope (DSO-X 2024A) had switchable input Z (50/75/1M) as I do loads of RF and video work.
I have to use T pieces and terms - very messy - easy to use the wrong term.

Using a spectrum analyzer and a tracking generator is a great way to show the effect of a stub.
Why would anyone put a 75 ohm terminator on a 50 ohm connector. There are 75 ohm BNCs, although they are rare these days. Beware of these things, as they cross-mate mate too easily with a 50 ohm BNC, and bust it.

Adding 50 ohm termination is not cheap if you do it right. You need a switch, and it probably needs to be a physical switch. If you switch in a 50 ohm terminator while the maximum permitted voltage is applied to the input there would be considerable power dissipated, so you also need protection too.

When the input switching of scopes was largely mechanical a 50 ohm terminator setting was much more common. Quite a few scope didn't have proper protection, though, and dealing with overdissipation in the terminator was a common reason for repairs.

When working with video you have to use a 75 ohm term. I use 75 ohm cable and a 75 ohm term plugged into a 50 ohm T-piece into the 'scope. The T-piece doesn't appear to suffered having the wrong BNC plugged into it. I would not risk a 75 ohm BNC plugged directly into the 'scope.

If you weren't meant to look at video then why give us all those video triggering modes?
Oscilloscopes are meant for the viewing of many types of signal, but the oscilloscope doesn't provide every possible interface needed for that.

If you have a real 75 ohm BNC it has a thinner pin in the middle, so its impedance is higher. They easily mate with a 50 ohm BNC, but the narrower socket will be stretched by the thicker pin. I have no idea when they didn't make the outside part of the 75 ohm version fatter to get the right impedance in a way that wouldn't allow false mating.

These days its common to find a 50 ohm BNC on the end of 75 ohm cables. YMMV.

Sorry,but that doesn't make sense!
If a 75 Ohm BNC centre pin was narrower than that of a 50 Ohm one, how could that damage the latter?
Perhaps the other way round!.

Over many years of dealing with 75 Ohm video systems in analog TV Broadcasting,I have found it was standard practice to use "50 Ohm" BNCs on the end of 75 Ohm coax--- they make them to fit RG59,etc.

With the advent of Digital TV,& the necessity of piping baseband digital video around the Studio,the racks were rewired with "real" 75 Ohm BNCs,as were any test leads,etc.

At the time,(1999),the only difference I could discern with the plugs was that the insulation on the 50 Ohm ones was brought right up to the front of the connector,whereas with the 75 Ohm one it was not.
There was never any question of incompatibility.

Looking at a manufacturer's site,unless I misread,the part number for the centre pin is the same for both versions,but the different insulation coverage is plain in the diagrams.

http://www.gordontech.com.au/catalog.htm

There is a definite difference in  centre pin size in "N" connectors,where the 75 Ohm one will cause damage to the 50 Ohm version,but that is a different matter.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 6388
  • Country: gb
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #36 on: August 19, 2014, 06:43:01 am »
Sorry,but that doesn't make sense!
If a 75 Ohm BNC centre pin was narrower than that of a 50 Ohm one, how could that damage the latter?
Perhaps the other way round!.

One way around you get poor contact. The other you get damage. Neither is a happy solution.

Over many years of dealing with 75 Ohm video systems in analog TV Broadcasting,I have found it was standard practice to use "50 Ohm" BNCs on the end of 75 Ohm coax--- they make them to fit RG59,etc.

With the advent of Digital TV,& the necessity of piping baseband digital video around the Studio,the racks were rewired with "real" 75 Ohm BNCs,as were any test leads,etc.

At the time,(1999),the only difference I could discern with the plugs was that the insulation on the 50 Ohm ones was brought right up to the front of the connector,whereas with the 75 Ohm one it was not.
There was never any question of incompatibility.

Looking at a manufacturer's site,unless I misread,the part number for the centre pin is the same for both versions,but the different insulation coverage is plain in the diagrams.

http://www.gordontech.com.au/catalog.htm

There is a definite difference in  centre pin size in "N" connectors,where the 75 Ohm one will cause damage to the 50 Ohm version,but that is a different matter.

I suspect the industry has completely abandoned the use of 75 ohm BNCs because of the cross mating problems. Analogue video isn't very wide bandwidth, and you kind of get away with having a small length of 50 ohm discontinuity in a 75 ohm cable. Try it at 300 MHz, and the results can be pretty sad.
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 5958
  • Country: au
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #37 on: August 19, 2014, 07:18:29 am »
Sorry,but that doesn't make sense!
If a 75 Ohm BNC centre pin was narrower than that of a 50 Ohm one, how could that damage the latter?
Perhaps the other way round!.

One way around you get poor contact. The other you get damage. Neither is a happy solution.

Over many years of dealing with 75 Ohm video systems in analog TV Broadcasting,I have found it was standard practice to use "50 Ohm" BNCs on the end of 75 Ohm coax--- they make them to fit RG59,etc.

With the advent of Digital TV,& the necessity of piping baseband digital video around the Studio,the racks were rewired with "real" 75 Ohm BNCs,as were any test leads,etc.

At the time,(1999),the only difference I could discern with the plugs was that the insulation on the 50 Ohm ones was brought right up to the front of the connector,whereas with the 75 Ohm one it was not.
There was never any question of incompatibility.

Looking at a manufacturer's site,unless I misread,the part number for the centre pin is the same for both versions,but the different insulation coverage is plain in the diagrams.

http://www.gordontech.com.au/catalog.htm

There is a definite difference in  centre pin size in "N" connectors,where the 75 Ohm one will cause damage to the 50 Ohm version,but that is a different matter.

I suspect the industry has completely abandoned the use of 75 ohm BNCs because of the cross mating problems. Analogue video isn't very wide bandwidth, and you kind of get away with having a small length of 50 ohm discontinuity in a 75 ohm cable. Try it at 300 MHz, and the results can be pretty sad.

The "Impedance discontinuity" isn't that great,& only occupies a very small fraction of a wavelength,even at 300MHz..
So-called "UHF" connectors which don't have any  declared Impedance "get away with it"at 438MHz!

Either there was an earlier version of 75 Ohm BNC which I am unaware of (& I've been around a long time),or the compatibility problem isn't as bad as you think.
 

Offline Slothie

  • Regular Contributor
  • *
  • Posts: 66
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #38 on: August 19, 2014, 11:53:12 am »
I have a small box full of T pieces and terminators somewhere because I'm old enough to have been working with computers when thinwire coax was the network cabling of choice :) This video shows WHY accidently disconnecting a cable could shut down the network for a whole section!! :palm:
 

Offline ConKbot

  • Super Contributor
  • ***
  • Posts: 1255
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #39 on: August 19, 2014, 04:31:13 pm »
If youre that concerned about it
http://www.minicircuits.com/MCLStore/ModelInfoDisplay?14084643973520.7602220219820596

50 ohm end goes into scope, 75 ohm end goes to your 75 ohm cable.  Scope has to be in 50 ohm mode, and just remember to account for the 5.7db of loss. 
 

Offline peteroakes

  • Contributor
  • Posts: 15
  • Country: ca
    • The Breadboard
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #40 on: August 20, 2014, 02:29:24 am »
 :-+ nice video Dave, many things I had not mentioned in my video when describing how use the same properties to measure the length of various cables and how the termination can mess you up my video for reference is here, http://youtu.be/JPnKfPzj3E4 , it is not talking about the pitfalls from a measurement perspective, I think you did an admirable job of that, but more about how to make use of those pitfalls to determine where a fault lies and also how to measure the length of a cable even if you dont have access to the other end of it.

I also only used a Sig gen and an Ebay special scope so as you said, its all about the fast rise times of the signal source and does not require anything too expensive to see the affect

Peter from the Breadboard.ca

Keep up the excellent work Dave.
 

Offline larry42

  • Regular Contributor
  • *
  • Posts: 173
  • Country: 00
Re: EEVblog #652 - Oscilloscope & Function Generator Measurement Trap
« Reply #41 on: August 28, 2014, 06:26:56 pm »
I wish my 'scope (DSO-X 2024A) had switchable input Z (50/75/1M) as I do loads of RF and video work.
I have to use T pieces and terms - very messy - easy to use the wrong term.

Using a spectrum analyzer and a tracking generator is a great way to show the effect of a stub.
Why would anyone put a 75 ohm terminator on a 50 ohm connector. There are 75 ohm BNCs, although they are rare these days. Beware of these things, as they cross-mate mate too easily with a 50 ohm BNC, and bust it.

Adding 50 ohm termination is not cheap if you do it right. You need a switch, and it probably needs to be a physical switch. If you switch in a 50 ohm terminator while the maximum permitted voltage is applied to the input there would be considerable power dissipated, so you also need protection too.

When the input switching of scopes was largely mechanical a 50 ohm terminator setting was much more common. Quite a few scope didn't have proper protection, though, and dealing with overdissipation in the terminator was a common reason for repairs.

I used to think so as well (damage between 50 and 75Ohm BNC), but this seems to *not* be the case for modern 75Ohm BNC connectors. I checked this for a project that used DIN1.0/2.3 and BNC connectors (both 50 and 75Ohm). According to Rosenberger
"The optimum working frequency of BNC and TNC 50 ? connectors isupto4GHz(max.10GHz),of75?connectorsisupto1GHz (max. 4 GHz). 50 ? and 75 ? connector types are intermateable without any restrictions within BNC and TNC series."
http://www.rosenberger.com/documents/headquarters_de_en/ba_communication/catalog_coax/12_Chapter_BNC_TNC.pdf

I believe that this change was made in the 70s or 80s.

NB this is *not* the case for 75Ohm N-type, but these are thankfully quite rare (I've never seen one).

If you have an animated GIF in your avatar or signature then I reserve the right to think you're a dolt.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf